Available online at www.sciencedirect.com Forest Ecology and Management ELSEVIER Forest Ecology and Management 185 (1003) 179-191 w~~.elsevier.comilocate~foreco

Options for managing early-successional forest and shrubland bird habitats in the northeastern Richard M. ~e~raaf",*,Mariko yamasakib "USDAForest Service. Northeastern Research Station. Universie of Massacln~setrs.Amherst. MA 01003. USA h~~~~ Forest Service. Northeastern Research Station. P.0. Box 640, Durham, NH 03824. USA

Abstract

Historically, forests in the northeastern United States were disturbed by fire, wind, Native American agriculture, flooding, and (Castor canadensis). Of these, wind and beavers are now the only sources of natural disturbance. Most disturbance- dependent species, especially birds, are declining throughout the region whereas species affiliated with mature forests are generally increasing or maintaining populations. Disturbance must be simulated for conservation of early-successional species, many of which are habitat specialists compared to those associated with mature forests. Both the maintenance of old fields and forest regeneration are needed to conserve brushland species. Regenerating forest habitats are more ephemeral than other woody early-successional habitats. The types and amounts of early-successional habitats created depend on the silvicultural system used, patch size selected, time between regeneration cuts, and rotation age. We recommend that group selection and patch cuts should be at least 0.8 ha, and patches should be generated approximately every 10-15 years depending on site quality. Kegeneration of intolerant and mid-tolerant tree species should be increased or maintained in managed stands. Also, frost pockets, unstocked, or poorly-stocked stands can provide opportunities to increase the proportion of early-successional habitats in managed forests. Published by Elsevier B.V.

Kew.ords: Disturbance-dependent species; Early-successional habitats: Even-age management: Opening size; Silvicultural systems

1. Introduction practiced mainly along the coast and major rivers. Natural disturbances were frequently a result of indi- Forests in the northeastern United States were vidual tree falls and larger blowdowns. Fire was a historically subject to several sources of disturbance. minor source of disturbance, but of local importance in In much of the region, early-successional habitats areas of glacial outwash sands and gravels, and on were continuously produced in pre-settlement times shallow soils at higher elevations (Lorimer and White, by fire, wind, (Castor canadensis), flooding, 2003). In much of the Northeast, many fire-prone areas and Native American agriculture and burning (see were occupied upon settlement by Europeans and are DeGraaf and Miller, 1996, pp. 6-10 for review). not available for management by prescribed fire. In northern areas, Native American agriculture was Historical levels of disturbance by beavers and flood- ing were in all likelihood significant and probably the main source of small meadows and shrubland habitats ' Corresponding author. Tel.: + 1-413-545-0357; fax: + 1-413-545-1 860. in much of the region (Askins, 2000, p. 12). Beavers E-mail address: [email protected] (R.M. DeGraaf) were extirpated and most streams and rivers dammed

0378-1 127/$ - see front matter. Published by Elsevier B.V doi: 10.1016/S0378-1127(03)00254-8 180 R.M. DeGraaf: M. Yamasaki/Forest Ecology and Management 185 (2003) 179-191

before their effects on the landscape were described. mid-1900s (McKinnon et al., 1935; Litvaitis, 1993), Beavers are now increasing in the region but cannot and a wave of early-successional habitats followed. modify the land as in presettlement times, so these Today, such habitats are less common than they were sources of disturbance are now greatly diminished in presettlernent times in several regions of the north- from historical levels. Wind still creates small open- eastern US, especially southern and south-coastal New ings in softwood stands, especially balsam (Abies England (Litvaitis, 1993; DeGraaf and Miller, 1996) balsamea) in mountainous areas, but young hardwood and in the coastal mid-Atlantic region (Lorimer and stands, now common across much of southern New White, 2003). On the other hand, the proportion of England, are fairly resistant to humcane damage early-successional habitat in northern industrial for- (Leak et al., 1994). Major humcanes occur at about ests is currently several times that which occurred in 150-year intervals (Channing, 1939). Considering all presettlement times (Lorimer and White, 2003). The intensities of blowdown, the average interval across northeastern US landscape is now dominated by the landscape is about 14 years in central New Eng- human uses, and maintaining early-successional habi- land (Leak et al., 1994). Such events are episodic and tats throughout in proportion to presettlement levels is cannot be relied upon to create disturbance where not possible. If the concern is for conservation (i.e. and when it is needed. As a result, management is maintenance of viable populations) of early-succes- needed to provide habitats for disturbance-dependent sional species, management needs to be representative species. of the "range of historical variation" where feasible. Recently, forest managers have been encouraged to Adherence to presettlement levels is not possible in consider baseline conditions when selecting a range of much of the northeastern US, and proposing manage- forest age distributions. All historical baselines are ment that does not create adequate patches of early- relative or arbitrary, however. Many different vegeta- successional habitats, while within the "natural range tive conditions existed during the last 12,000 years in of variation" locally, will tend to eventually eliminate post-glacial New England: parkland, grass- such habitats and the species associated with them lands and open forest maintained by large browsing throughout the region. ungulates and mastodons (Askins, 2000, p. 17), open Most wildlife associated with native shrublands and areas created by Native American agriculture and early-successional habitats were once considered gen- burning, dense forests after Native American retreats eralist species that flourished at edges and needed no and declines, the extensively cleared landscape pro- specific management actions. Many early-succes- duced by an agrarian society following European sional taxa, however, are now either extinct, extir- settlement, and the present largely forested condition pated, threatened, or species of management concern (DeGraaf and Miller, 1996). Choosing the conditions in the region, largely because of habitat loss (Witham that existed just prior to European settlement is arbi- and Hunter, 1992; Litvaitis, 1993). Most early-succes- trary and is no more indicative of a "natural" land- sional species are not generalist species at all; rather, scape condition than any of the others. Given the wide they are, to a remarkable degree, specialists in vegeta- historical range of vegetative conditions that have tion structure or area requirements. Only their former existed in the post-glacial era (Lorimer, 2001), mana- abundance led to their erroneous labeling. For exam- ging forested landscapes within the "natural range of ple, New England cottontails (Sylvilagus rmnsitiona- variability" has been proposed for maintaining the lis) are obligate users of young stands, occupying conditions to which wildlife species have become habitats when secondary succession has progressed adopted (Seymour and Hunter, 1999; Thompson 10-25 years, after which time they decline rapidly and DeGraaf, 2001). when understories thin as trees mature (Litvaitis, Early-successional wildlife habitats have now 2001; Barbour and Litvaitis, 1993). New England become critically uncommon in much of the eastern cottontails are also area-sensitive; individuals occupy- United States (Trani et al., 2001), especially the North- ing small (<3 ha) patches of habitat encounter winter east (Brooks, 2003), largely in response to forest food shortages (Villafuerte et al., 1997). Cottontails maturation. Abandonment of agricultural lands that forage away from escape cover are killed by reached a peak in New England in the late 1800s to predators at about twice the rate as cottontails on R.M. DeGraaf: M. Yamaruki/Forest Ecology and Manogernettt 185 (2003) 179-191 181 large patches (Barbour and Litvaitis, 1993; Villafuerte fasciarus), Swainson's thrush (Carharus usrulatus), et al.. 1997). -sided warblers (Dendroica magnolia warbler (Dendroica magnolia). Cape May pensylvanica) generally occupy patches of regenerat- warbler (Dendroica tigrina), black-throated blue war- ing hardwoods at least 0.2 ha in New England (King bler (Dendroica caerulescens), and worm-eating war- et al., 2001). and then only for 3-10 years after bler (Helmitheros vermivorus)]. The majority of harvest. Chestnut-sided warblers build well-concealed vertebrates in New England are quite plastic in their nests 0.3-1.2 m above ground in low bushes or sap- habitat distributions (DeGraaf et al., 1992; DeGraaf lings (DeGraaf and Yamasaki, 2001, p. 221), and and Yamasaki. 2001). Early-successional habitats are decline steadily in a patch as stand development in general decline (Thompson and DeGraaf, 2001). reduces the availability of nesting habitat. While some species depend on agriculture or frequent Grassland and shrubland birds as a group best burning or mowing (Oehler, 2003). the greatest oppor- illustrate the habitat specificity of early-successional tunity to provide habitat for most species lies in even- species compared to that of mature forest species age forest management (DeGraaf et al., 1992, p. 131). (Askins, 1993). For example, Henslow's sparrows In this paper, we review the important habitat (Ammodramus henslowii) occur only in fields with a characteristics of early-successional habitats in the deep litter layer, standing dead forbs and tall. dense northeastern United States and suggest management grass (Zimmerman, 1988). Yellow-breasted chats options for increasing the amount and quality of these (Icteria virens) inhabit brushy old fields only until habitats. Although our comments and recommenda- they begin to be invaded by overtopping trees (Shugart tions are directed toward management of avian habi- and James, 1973; Thompson, 1977). The decline of tats, we believe the principles covered will have agriculture and reversion to forest have essentially application to other taxa dependent on early-succes- eliminated grassland birds from the New England sional habitats. landscape, although many were common a half-cen- tury ago (Bagg and Eliot, 1937; Askins, 1993). - land birds that are not adapted to suburbia, such as 2. Natural shrublands and regeneration brown thrashers (Toxostoma rufum), eastern towhees habitats (Pipilo erythrophthalmus), and field sparrows (Spi- :ella pusilla), are now declining across the region Natural shrublands are among the most endangered (Veit and Petersen, 1993; DeGraaf and Yamasaki, ecosystems in the United States (Noss et al., 1995). 2001 ). Mature forest birds, whose habitats are increas- Woody, early-successional communities are domi- ing in eastern North America, are as a group quite nated by , young trees, and, to varying degrees, tolerant of dramatic changes in stand structure (Webb by grasses and forbs. In the northeastern US, nearly all et al., 1977) or patchy disturbance within areas of shrub communities are successional. The most stable extensive forest (Maurer et al., 1981; DeGraaf, 1991; shrub communities occupy a range of sites, including Thompson et al., 1992). Grassland and shrubland hydric (Cowardin et al., 1979). sandy sites birds, however, are specialists that quickly disappear (Patterson and Sassaman, 1988), and ledge areas from a site as succession proceeds (Litvaitis, 1993). (Kimball et al., 1995). In the Northeast, shrub-domi- In New England, approximately 200 vertebrate nated communities persist the longest in extremely dry species occur in shrublold field habitat and regenerat- or wet areas, at high elevations, and places exposed to ing stands of aspen, paper birch, northern hardwoods, salt spray (Latham, 2003). The character of the woody balsam fir, spruce-fir, and -pine types in New vegetation that develops on a site varies with type of England (DeGraaf and Yamasaki, 2001). Relatively disturbance. For example, old-field succession that few species in the region [e.g. northern hanier (Circus follows agricultural abandonment is a result of the cyaneus), savannah sparrow (Passerculus sandwichen- invasion of woody plants from the surrounding land- sis), and vesper sparrow (Pooecetes gramineus)] are scape. It takes much longer for an old field to be limited to non-forest habitat types. There also are reoccupied by trees than for a regenerating forest stand relatively few mature forest species that also occur to become re-established (Tryon, 1945; Stephens and in regenerating forest [e.g. five-lined skink (Eumeces Ward, 1992). Succession is most rapid on forest sites 182 R.M. DeGraaJ M. Yamosaki/Forest Ecology and Managrmenr 185 (2003) 179-191 where tree seedlings and sprouts grow rapidly follow- Early-successional habitats created by timber har- ing a severe canopy disturbance. vests are dominated by tree reproduction and differ In present-day landscapes, more frequent distur- from other early-successional habitats. A useful dis- bances and large gaps (e.g. due to fires and hurricanes) tinction between "early-successional habitat," which occur among coastal and valley bottom forests and on is dominated by herbs, shrubs, and pioneer tree species mountains and hilltops. Less frequent disturbances such as aspen or pine, and "young forest habitat," and small gaps (eg due to blowdown, 'insect infesta- which includes young stands of late successional tion, and ice storms) occur in upland and midslope species such as sugar maple (Acer saccharurn) has forests (Lorimer, 2001; Lorimer and White. 2003). been proposed by Lorimer (200 I). Also, some shrub- Land abandoned by beavers is recolonized by grasses, land habitats may be more patchy by nature, and forbs, sedges, shrubs, and trees, and historically contain more vines and shrubs than do regenerating occurred more frequently on lower slopes and along forest stands, and may have some unique species. low-gradient streams than on mid- and upper slope Shrublands in the northeastern US also contain high positions (Hodgdon and Hunt, 1953; Howard and proportions of rare plants. Rich examples exist in the Larson, 1985). Barrens and other xeric shrublands White Mountains of New Hampshire, where 15 of 62 are maintained in an arrested state of succession by rare plants are shrubs; Maine and New Jersey also edaphic factors and fire (Tubbs and Verme, 1972; contain globally rare shrublands (Latham, 2003). Patterson and Backman, 1988; Whitney, 1994). While forest regeneration and shrublands provide

Table I Available vegetative characteristics of early-successional habitats in northeastern United States

Vegetative characteristics Bums" Old fieldsh Regenerating Regenerating hardwoods HBEF hardwoods BEF'.' Overstory BA m2 >I0 cm dbh variable' variable' Nil <0.93 Trees

Percent total stems (%biomass) (10' stemsha)

Shrub layer (1-3 m tall) I -year post-treatment - - 8. 19 - 5-year post-treatment - - 24. 4 16.641 10-year post-treatment - - 8. 2 -

Percent total stems (%biomasb) Percent coverages

Grasslforb layer (

"Mills (1961). Olson ( 1965). ' Martin and Hornbeck (1989). ' Costello et al. (2000). ' DeGraaf and Yamasaki (2002) Stephens and Ward (1992). "amasaki (unpublished data). R.M. DeGroof; M. Yarnaraki/Forest Ecolog,~ habitats for many of the same species, both are needed Table 2 to provide habitats for the full range of disturbance- Species composition 1C-15 years after cutting in -birch- maple type by tolerance group and cutting method (from Leak and dependent species (Thompson and DeGraaf, 2001 ). Wilson. 1958) Although data are limited, the structure of different types of woody early-successional habitats in New Tolerance Clearcut Group selection Individual tree England (e.g. bums, old fields, and regenerating for- category" (%) (54) selection (%) ests) seem to be different at the start (Table 1). Tolerants 43 62 92 Grasses, herbs, and shrubs persist longer on burns Intennediates 19 34 7 and old fields than in regenerating forest habitats. Intolerants 38 4 I Young forest habitats are therefore more ephemeral. "Species included in each category. Tolerants: beech (Fagus For example, the seedling stage of northern hardwood gra~~d+dicr).sugar maple (Acer sacclrarurr~), eastern hemlock clearcuts commonly last about a decade (Leak et al., (Tsrcga cu~~aiier~sis),and red spruce (Picea rrrberrs). Intermediates: yellow birch (Betula olleghnnier~sis).white ash (Fraxinus amer- 1987). Among old fields, this stage commonly persists icnrlo). and red maple (Acer nrbnrnt). Intolerants: aspen (Populus for two to three decades (Stephens and Ward, 1992), spp.) and paper birch (Bctula paprifera). and some native shrub communities can persist two to three centuries (Latham, 2003). between regeneration cuts, and rotation age or re-entry The growth of a forest stand after a major distur- period. Silvicultural prescriptions designed to gener- bance is characterized by four developmental stages: ate early-successional habitats utilize relatively large stand initiation, stem exclusion, understory reinitia- regeneration cuts that create favorable environments tion, and old growth (Oliver and Larson, 1996). These for dense reproduction of intolerant, intermediate, and stages are commonly recognized as seedlinglsapling, tolerant tree species (Leak and Wilson, 1958; Marquis, poletimber, sawtimber, and large-sawtimber stands. 1967; Smith et al., 1997). Aspen (Populus spp.), birch (Berula spp.). and north- The extent that the canopy is removed influences the em hardwood forests may remain in the stand-initia- species composition of the developing stand (Table 2). tion stage for

Also, the movements of post-fledgling juveniles likely eri), eastern bluebird (Sialia sialis), and mourning reflect foraging optimization; they need abundant food warbler (Oporomis philadelphia), will therefore be resources to complete the prebasic molt and accumu- habitat-limited where only single-tree and small group late fat reserves for migration. For example, juvenile selection practices are used. Single-tree selection will wood thrushes have been shown to leave nearby not create early-successional habitat under any cir- mature forest habitat and enter early-successional cumstances. habitat to forage on fruits and invertebrates (Anders Forest stand edge effects in forest-dominated land- et al., 1998; Vega Rivera et al., 1998). scapes in the northern United States are ephemeral Effect of opening size in forest-dominated land- because woody regeneration in cut stands grows scapes also influences avian species richness (Taylor quickly (DeGraaf and Yamasaki, 2002). Distinct and Taylor, 1979). Rudnicky and Hunter (1993a) breeding bird assemblages do not occur at forest reported that breeding bird species richness increased interior stand edges; rather, distinct avian assemblages in regenerating spmce-fir stands and clearcut areas up are separated by abrupt interior edges [e.g. seedling1 to 20 ha. Costello et al. (2000) compared songbird sapling-sawtimber stand interfaces (DeGraaf, 1992)l. response to group selection cuts (0.13-0.65 ha) to Evidence from studies using artificial nests shows clearcuts (8-12 ha) and found that species richness variable effects on edge-related depending for birds associated with open habitats was higher on location, egg and nest types used, and other factors in clearcut openings than group selection openings, (see reviews by Paton, 1994; Major and Kendal, 1996). suggesting that smaller group selection cuts do not Edge-related increases in nest predation are well provide the equivalent habitat for breeding birds that documented from both studies using artificial nests similarly aged larger clearcuts do. King et al. (2001) (Rudnicky and Hunter, 1993b) and natural nests (King found that daily nest survival rates for early-succes- et al., 1996, 1998). Studies in extensive forests in the sional birds were similar in large group cuts and Northeast show that silvicultural edges do not result in clearcuts in extensive northern hardwood forests. elevated nest predation rates (DeGraaf and Angelstam, Early-successional bird species that need large forest 1993; DeGraaf, 1995; VanderHaegen and DeGraaf, gaps, such as olive-sided flycatcher (Contopus coop- 1996; DeGraaf et al., 1999) in contrast to suburban

Table 3 Number of years after clearcutting an eastern forest that breeding, early-successional birds first appear, become common, and then decline Bird species First appear Become common Decline

Ruffed grouse (Bonasa urnbellus) Northern flicker (Colaptes auratus) Olive-sided flycatcher (Contopus cooperi) Willow flycatcher (Empidonax traillii) Tree swallow (Tachycineta bicolor) Winter wren (Troglodytes troglodytes) Eastern bluebird (Sialia sialis) Veery (Catharus fiscescens) Swainson's thrush (Catharus ustulatus) Cedar waxwing (Bombycilla cedmrum) Chestnut-sided warbler (Dendmica pensvlvanica) Black-and-white warbler (iMniorilta varia) Mourning warbler (Opommis philadelphia) Common yellowthroat (Geothlypis rrichas) Canada warbler (Wilsonia canadensis) White-throated sparrow (Zonotrichia albicollis) Rose-breasted grosbeak (Pherrcticus ludovicianus~

We assumed that some residual stems (snags and live trees) remain (DeGraaf. 1987. Thompson et al., 1992. Thompson et al.. 1996) "Present until next cutting cycle. R.M. DeGraaf: M. Yamasaki/Forest Ecology and Martagrrncrlr 185 (2003) 179-191 185 forest edges (Danielson et al., 1997). King et al. (1998) vicultural practices that minimize the size of regen- found that forest edge-related nest predation was eration cuts and the total harvest area or lengthen similar for both clearcut and group selection stands rotation age and re-entry periods inhibit the creation in forest-dominated landscapes in New Hampshire. It and maintenance of young forest habitats. Early-suc- would seem that in extensively forested landscapes in cessional bird species are capable of locating and northern areas, edges produced by even-age manage- using widely dispersed disturbance gaps throughout ment does not result in an increase in nest predation a northeastern landscape as these habitats become rates. available (Lent and Capen, 1995). It has been sug- Frequency of disturbance also matters because gested that natural disturbance gaps would provide a breeding bird composition changes rapidly in the first base level of sufficient habitat. Now that historical 10-15 years after complete clearcutting. Many of the sources of disturbance are much reduced, we speculate earliest arriving birds decline in just a few years as that small, infrequent, and isolated regeneration cuts in habitat conditions change (Table 3). In the White a large watershed, whether created by clearcut, patch, Mountains of New Hampshire and Maine, regenerat- or large group selection methods, would take a long ing stands 1-5 years old contain about 28 bird species. time to be occupied by these species, given the lack of Of these, five are restricted to that stage. Sapling nearby source populations. Avian occupation of iso- stands contain about 30 species and pole-timber stands lated patch cuts (1.2-2.4 ha) at the Bartlett Experiment only about half as many (DeGraaf, 1987). White- Forest in central New Hampshire, where the percentage throated sparrows (Zonotrichia albicollis), winter of regenerating habitat in the watershed is low (<3%) in wrens (Troglodytes troglodytes), and willow flycatch- any given decade, seems to occur at a slower rate than ers (Empidonax traillii) are generally abundant in the in larger clearcuts in more actively managed water- first growing season after complete removal of all live sheds (Costello, personal communication). stems. Winter wrens are associated with dense slash. If In southern New England, where large openings or stubs with old woodpecker holes are left, eastern clearcuts are socially unacceptable, group selection is bluebirds are commonly present. Two years after the most likely method to create suitable habitat. clearcutting, the number of species doubles. Common Groups >0.5 ha need to be cut and clustered closely yellowthroats (Geothlypis trichas), chestnut-sided together to maintain an adequate level of early-succes- warblers, cedar waxwings (Bombycilla cedrorum), sional habitat. Strategies suggested by Hagan et al. American goldfinches (Carduelis tristis), and mourn- (1997) and King et al. (2001) to consolidate both ing warblers will invade, along with Swainson's regenerating and mature forest stands into larger thrushes, eastern towhees, and American redstarts blocks would increase the likelihood of systematically (Setophaga ruticilla). Northern flickers (Colaptes aur- and effectively providing habitats for both mature atus) and white-throated sparrows remain, but eastern forest and early-successional habitat bird species. bluebirds and sometimes winter wrens are gone. In the third growing season after clearcutting, bird species numbers again double, with about a dozen new species 4. Management of shrublands in a forest added, including black-and-white warbler (Mniotilta landscape varia), rose-breasted grosbeak (Pheucticus ludovicia- nus), mostly in low numbers. During the next 12 years, Several opportunities exist to provide or maintain bird species composition changes substantially, as shrublands in otherwise forested landscapes. Natural ruffed grouse, red-eyed vireos (Vireo olivaceus), wood opportunities exist in frost pockets and on poor or thrushes (Hylocichla mustelina), and ovenbirds depleted sites (Tubbs and Verme, 1972). Old fields, (Seiurus aurocapillus), among others, arrive but the depending on soil conditions, seed bank (Leck and number of species usually does not change appreci- Leck, 1998), density of herbaceous growth and seed- ably (DeGraaf, 1991). ling mortality (Gill and Marks, 1991), and densities of The time between periodic regeneration cuts needs herbivores (Manson et al., 2001) and deer to be short enough to maintain the presence of early- (Inouye et al., 1994) can provide early-successional successional bird species in a management area. Sil- habitat for 30-50 years. 186 R.M. DeGraaf; M. Yotnasaki/Forc..~~Ecologv and Mana,qetnent 185 (2003) 179-191

Powerline rights-of-way can contribute significant source of shrublands and early-successional forest shrubland habitat area. For example, a 100 m wide habitat in the first 10-15 years post-clearcut. The corridor provides 1 ha of shrubland habitat for each percentage of early-successional woodcock habitat 100 m it extends through forest. Powerlines in the in landscapes (>25%) proposed by McAuley et al. Northeast constitute a base on which to build useful (1996), can most easily be met on industrial forest core shrubland habitat in human-dominated land- landscapes. Site preparation and stand culture activ- scapes where natural disturbance patterns are obviated ities often involve herbicide application. targeting by landscape fragmentation. The New England cotton- hardwood competition to release to grow tail, proposed for listing as an endangered species, quickly (Seymour et al., 1986). Current herbicide does not persist in small patches in human-dominated formulations temporarily reduce hardwood browse, landscapes. This species would likely benefit most but re-sprouting and herbicide-resistant hardwoods from the provision of 'core habitat' patches at least and untreated 'skip' areas often result in dense shrub 10 ha in size, clustered within several kilometers of patches in which numerous shrub-dwelling birds and each other (Litvaitis, 2001) and positioned along browsing can be found (e.g. Hagan et al., adjacent early-successional habitats such as powerline 1997; Raymond et al., 1996; Santillo et al., 1989). corridors (Askins, 1993). Powerline corridors also Precommercial thinning and spacing activities remove provide suitable shrubland bird habitat. For example, the nearest hardwood competition from the desired nesting and fledging success of chestnut-sided war- stems, and still maintain dense hardwood blers in western Massachusetts powerline corridors regeneration and shrub zones until the conifer overs- were high, and reproductive success and adult survival tory closes. also creating productive woodcock habitat were sufficient to balance losses from mortality (King (McAuley, personal communication). These habitat and Byers, 2002). Also, nesting success near utility elements, when combined with the extensive rights-of-way in heavily-forested landscapes in New (Alnus sp.) acreage, can form high value shrubland York, Massachusetts, and Maine was higher than that habitat across the industrial forest landscape during in the rights-of-way and similar to that in forest, the period prior to overstory crown closure. suggesting that negative effects of edge observed in agricultural landscapes are not associated with rights- of-way through forested landscapes (Confer and Pas- 5. Practical approaches for providing coe, 2003). early-successional habitats Intensive high-yield in the industrial forest landscapes of northern New England (Seymour We suggest using habitat composition goals and Hunter. 1992) also can provide an ephemeral to maintain a balanced and integrated set of forest

Table 4 Structural characteristics commonly found in regenerating hardwood and softwood forests in the northeastern United States Structural characteristics clearcup.' ~helterwood~ Group ~election".~.".".' Single-tree selection^“ Average gap size (ha) >8.1 Variable 0.2 0.002-0.03 Size range of gaps (ha) 4-30.4 Variable 0.04-0.81 0.002-0.03 Likelihood of regenerating intolerants Moderate-high Moderate Low-moderate None and mid-tolerant tree species Time until next stand entry (years) 4MO 5660 15-30

" Leak (1983. 1999). Leak et al. (1 987). ' Marquis (1967). McClure and Lee (1993). McCIure el al. (2000). ' Lee (personal communication) "imball et al. (1995). Seymour et al. (2002). R.M. DeGraaf; M. Yar~tasaki/Forest Ecology and Martagentolt 185 (2003) 179-191

I 0-year entrj period wildlife approach

20- year entry period traditional timber approach .. 20 0.3-ha culs 4 ?-ha cuts 1 8.1-ha cut

Increasing cut size and earl)-successional hab~la~quality j

Cut in year 1 on a 10-yr ently I Cut In year I on a 20-yr entry Cut in year I I on a 10-yr entry Assumes 40.5 ha; I00 !ear rotation

Fig. I. Comparison of stand entry periods under sustainable, regulated even-age management for providing continuous early-successional wildlife habitat vs. traditional silviculture in the northeastern United States. conditions that includes early-successional habitats intolerant and mid-tolerant tree regeneration in man- and young forest as well as mature and old forest aged stands; and (5) consider frost pockets and for a broad diversity of species over time (e.g. DeGraaf unstocked or poorly-stocked stands as opportunities et al., 1992). Optimizing early-successional habitats in to maintain and increase the area of early-successional northeastern landscapes would entail modifying these habitat. composition objectives to reflect higher proportions of We suggest applying these recommendations at regenerating forest and perhaps shorter rotational both the landscape and stand level to improve the cover-types; the structural composition of the result- quality and quantity of early-successional habitats in ing habitats would vary with silvicultural treatment the northeastern US. Landscape considerations (Table 4). Such management would likely not impact involve the regenerating forest acreage and permanent late-successional forest species if disturbance patches openings (e.g. powerlines and gas pipeline rights-of- were clustered and large blocks set aside for manage- way), brushy post-agricultural set-aside acreage, and ment on long (>I25 years) rotations. scrub-shrub wetlands. Together these habitat types Where clearcutting is not an option, managers can should comprise between 10 and 20 percent of the use the example in Fig. 1 to modify traditional timber forest landscape to optimize early-successional spe- management activities in several ways: (1) maintain cies diversity (DeGraaf et al., 1992). Three principles even-age management systems in the mix of forest based on the literature review above should guide management practices; (2) where using single-tree management where early-successional habitats are selection systems exclusively across the landscape, needed to arrest or reverse species declines: increase the use of group selection and patch cut methods to increase the gap size to at least 0.8 ha; Size of opening affects both woody intolerant (3) consider splitting a re-entry period by cutting half species composition in the regenerating stand and of the allowable volume in the first 10-15 years; the breeding bird species richness in the regenerating second half of the allowable volume per re-entry stand. Small (<0.25 ha) group selection cuts do period in years 20-25; (4) maintain or increase the not provide the equivalent habitat for breeding R.M. DeGraaf: M. Yamasaki/Forest Ecolop and Managentent 185 (2003) 179-191

birds that similarly aged larger clearcut units do. openings, shrub swamp wetlands, the O-10-year age Avian reproductive success is similar for species class and shorter rotation cover-types (intolerants); (b) nesting in clearcuts and large group cuts. a variety of opening sizes >0.5 ha; (c) a frequent (i.e. 2. Frequency of disturbance matters; the bird com- -10 years) re-entry period in places; and (d) cluster- munity in young hardwood forest is very ephem- ing these types of activities near existing and main- eral in the first 10 years after harvest and differs tained permanent openings-old fields, powerlines, almost completely from those occupying mature frost pockets, old burns, and predictably non-stocked stands. Edge effects in extensively forested land- sites. In much of the northeastern United States, scapes are minimal and very different from those especially along the coast and in the major river in forest-field landscapes. Cowbird occurrence and valleys, human population growth and development thus brood parasitism is very low in extensively continue to increase. These areas were once frequently forested New England landscapes (Yamasaki et al., disturbed by Native American agriculture, fire, and 2000) compared with forest-field landscapes hurricanes and have high potential species richness (Coker and Capen, 2000). (Askins, 2000; DeGraaf and Yamasaki, 2001). The 3. Spatially and temporally isolated patches of early- continued occurrence of early-successional species successional habitat in extensively forested land- depends upon both managing forests to the extent scapes likely have lower rates of occupancy for possible in historically-disturbed valley bottoms, invertebrates such as the Karner blue butterfly and to simulate disturbance to create such habitats (Lycaeides melissa samuelis) as well as nesting in uplands and on lower mountain slopes. Such actions [e.g. painted (Chrysemys p. picta), snap- would likely increase the rate of disturbance some- ping (Chelydra s. serpentina), and Blanding's what from historical levels of natural disturbance in (Emydoidea blandingii)], snakes [e.g. milk (Lam- upland forests, but rates would likely not exceed propeltis t. triangulum), black racer (Coluber c. historical levels for the region as a whole. Of one constrictor), and smooth green (Liochlorophis thing we can be sure: continuous management will be vernalis)], and passerines [e.g. towhees, brown needed to conserve and manage these habitats now and thrashers, indigo buntings (Passerina cyanea), into the future or they and the species they support will bluebirds, mourning and chestnut-sided warblers]. be lost in much of the region. Management needs to Additionally, raptors would benefit from represent. to the extent possible, the historical range of sites [e.g. American kestrels (Falco sparverius), variation in habitat conditions, and not mimic only the red-tailed hawks (Buteo jamaicensis), and sharp- forested parts of the historical range in each local area. shinned hawks (Accipiter striatus)] and nesting sites [e.g. hamers and Cooper's hawks (Accipiter cooperii)]. A variety of mammals would benefit Acknowledgements from additional foraging habitat [e.g. , mea- dow (Microtus pennsylvanicus), meadow We thank William Leak, David King, John Scanlon, jumping mice (Zapus hudsonius), red fox (Vulpes Stephen DeStefano, James Oehler, Robert Askins, vulpes), gray fox (Urocyon cinereoargenteus), Frank Thompson, and John Lanier for their critical bobcat (Lynx rufus), white-tailed deer (Odocoileus reviews and Mary A. Strong for typing the manuscript. virginianus) and moose (Alces alces)]. In other words, there is some critical minimum habitat area required in time and space for each species to be References present, and more for them to exist as viable populations. While the actual threshold area for Anders. A.D., Faaborg, J., Thompson 111, F.R.. 1998. Postfledgin: many species is unknown, declining habitat trends dispersal. habitat use, and home-range size of juvenile Wood are working against their conservation. Thrushes. Auk 115, 349-358. Askins, R..A.. 1993. Population trends in grassland, shrubland, and forest birds in eastern North America. Curr. Omith. 11. 1-34, Managing these habitats then requires: (a) a sig- Askins, R..\.. 2000. Restoring North America's Birds: Lessons from nificant percent of the landscape in permanent forest Landscape Ecology. Yale University Press. New Haven. CT. R.M. DeGraaf; M. Yamasakr/Forest Ecology and Management 185 (20011) 179-191 189

Bagg, A.C.. Eliot. S.A., 1937. Birds of the Connecticut Valley DeGraaf, R.M.. Yamasaki. M., 2002. Effects of edge contrast on in Massachusetts. The Hampshire Bookshop, Northampton. redback distribution in even-aged northern hard- MA. woods. For. Sci. 48. 351-363. Barbour, MS., Litvaitis, J.A., 1993. Niche dimensions of New DeGraaf, R.M., Yamasaki, M., Leak, W.B., Lanier, J.W., 1992. England cottontails in relation to habitat patch size. Oecologia New England Wildlife: Management of Forested Habitats. 95, 321-327. General Technical Report NE-144. USDA Forest Service. Brooks, R.T.. 2003. Abundance, distribution, trends, and ownership Northeastern Forest Experiment Station. Radnor, PA, 271 pp. patterns of early-successional forests and native shrublands in DeGraaf, R.M.. Maier, T.J., Fuller. T.K.. 1999. Predation of small the northeastern U.S. For. Ecol. Manage. 185, 65-74. eggs in artificial nests: effects of nest position. edge, and Channing. W., 1939. New England Hurricanes 1635-1 815-1938. potential predator abundance in extensive forest. Wilson Bull. Walter Chaning, Inc.. Boston, MA. l l 1, 236-242. Coker, D.R.. Capen. D.E., 2000. Distribution and habitat associa- Dessecker. D.R., McAuley, D.G., 2001. Importance of early- tions of brown-headed cowbirds in the Green Mountains of successional habitat to ruffed grouse and American woodcock. Vermont. In: Smith, J.N.M., Cook, T.L., Rothstein, S.I.. Wildl. Soc. Bull. 29, 456-465. Robinson. S.K.. Sealy. S.G. (Eds.), Ecology and Management Gill, D.S., Marks, P.L., 1991. Tree and shrub seedling coloniza- of Cowbirds and their Hosts. University of Texas Press, Austin, tion of old fields in central . Ecol. Monogr. 61, 183- TX, pp. 236-243. 205. Confer, J.L.. Pascoe, S.M., 2003. Avian communities on utility Hagan 111, J.M.. McKinley, P.S., Meehan. A.L., Grove, S.L., 1997. rights-of-ways and other managed shrublands in the north- Diversity and abundance of landbirds in a northeastern eastern United States. For. Ecol. Manage. 185. 193-205. industrial forest. J. Wildl. Manage. 61, 718-735. Costello, C.A., Yamasaki, M., Pekins, P.J., Leak, W.B.. Neefus. Hodgdon. K.W., Hunt. J.H.. 1953. Beaver management in Maine. C.D., 2000. Songbird response to group selection harvests and Game Div. Bull. 3. Maine Department of Inland Fisheries and clearcuts in a New Hampshire northern hardwood forest. For. Game. Augusta. Ecol. Manage. 127, 41-54. Howard, R.J.. Larson, J.S., 1985. A stream habitat classification Cowardin, L.M., Carter, V., Golet, F.C., LaRoe 111, E.T., 1979. system for beaver. J. Wildl. Manage. 49, 19-25. Classification of wetlands and deepwater habitats in the United Inouye. R.S.. Allison, T.D., Johnson, N.C., 1994. Old field States. FWS/OBS-79/31. USDI Fish and Wildlife Service, succession on a Minnesota sand plain--effects of deer and Office of Biological Services, Washington, DC. other factors on invasion by trees. Bull. Torrey Bot. Soc. 121, Danielson, W., DeGraaf, R.M., Fuller, T.K., 1997. Rural and 266-276. suburban forest edges: effect on egg predators and nest Kimball, A.J., Witham, J.W.. Rudnicky, J.L., White. AS., Hunter predation rates. Landscape Urban Plan. 38, 25-36. Jr., M.L., 1995. Harvest-created and natural canopy gaps in an DeGraaf, R.M., 1987. Managing northern hardwoods for breeding oak-pine forest in Maine. Bull. Torrey Bot. Club 122. 115-123. birds. In: Nyland, R.D. (Ed.). Managing Northern Hardwoods. King, D.I., Byers, B.E.. 2002. An evaluation of powerline rights-of- Miscellaneous Publication 13. College of En\konmental way as habitat for early-successional shrubland birds. Wildl. Science and , State University of New York. Syracuse. Soc. Bull. 30. 868-874. NY. pp. 348-362. King, D.I., Griffin, C.R., DeGraaf, R.M., 1996. Effects of DeGraaf, R.M., 1991. Breeding bird assemblages in managed clearcutting on habitat use and reproductive success of the northern hardwood forests in New England. In: Rodiek, J., ovenbird in forested landscapes. Cons. Biol. 10, 1380-1386. Bolen, E. (Eds.), Wildlife and Habitats in Managed Land- King, D.I., DeGraaf, R.M., Griffin. C.R., 1998. Edge-related nest scapes. Island Press. Washington, DC, pp. 153-1 7 1. predation in clearcut and groupcut stands. Cons. Biol. 12. DeGraaf, R.M.. 1992. Effects of even-aged management on forest 1412-1415. birds at northern hardwood stand interfaces. For. Ecol. Manage. King, D.I., DeGraaf, R.M., Griffin, C.R., 2001. Productivity of 46, 95-1 10. early-successional shrubland birds in clearcuts and groupcuts in DeGraaf, R.M., 1995. Nest predation rates in managed and an eastern deciduous forest. J. Wildl. Manage. 65, 345-350. reserved extensive northern hardwood forests. For. Ecol. Latham, R.E., 2003. Shrubland longevity and rare plant species in Manage. 79, 227-234. the Northeast (USA). For. Ecol. Manage. 185, 21-39. DeGraaf, R.M., Angelstam, P., 1993. Effects of timber size-class on Leak, W.B., 1983. Stocking, growth, and habitat relations in New predation of artificial nests in extensive forest. For. Ecol. Hampshire hardwoods. Research Paper NE-523. Manage. 61, 127-1 36. Leak. W.B., 1999. Species composition and structure of a northern DeGraaf, R.M., Miller, R.I., 1996. The importance of disturbance hardwoods stand after 61 years of grouplpatch selection. North and land-use history in New England: implications for forested J. Appl. For. 16, 151-153. landscapes and wildlife conservation. In: DeGraaf, R.M., Leck, M.A., Leck, C.F., 1998. A ten-year seed bank study of old Miller, R.I. (Eds.), Conservation of Faunal Diversity in Forested field succession in central New Jersey. J. Torrey Bot. Soc. 125, Landscapes. Chapman and Hall, London, pp. 3-35. 11-32. DeGraaf, R.M., Yamasaki, M., 2001. New England Wildlife: Leak, W.B., Wilson Jt, R.W., 1958. Regeneration after cutting of Hahitat, Natural History and Distribution. University Press of old-growth northern hardwoods in New Hampshire. Station New England, Hanover, NH. Paper 103. USDA Forest Service. 190 R.M. DeGroaf. M. Yomasrrki/Forest Ecology and Managrntet~tI85 (2003) 179-191

Leak, W.B., Solomon. D.S., DeBald. P.S., 1987. Silvicultural guide Oehler, J.D.. 1003. State efforts to promote early-successional for northern hardwood types in the Northeast (revised). habitats on public and private lands in the northeastern United Research Paper NE-603. USDA Forest Service. States. For. Ecol. Manage. 185, 169-177. Leak, W.B., Yamasaki. M.. Smith. M.L.. Funk, D.T., 1994. Oliver, C.D.. Larson, B.C., 1996. Forest Stand Dynamics, second Selection criteria for forested natural areas in New England, ed. Wiley, New York. USA. Nat. Areas J. 14, 300-305. Olson, A.R.. 1965. Natural changes in some Connecticut wood- Lent, R.A., Capen. D.E.. 1995. Effects of small-scale habitat lands during 30 years. Bulletin 669. CT Agricultural Experi- disturbance on the ecology of breeding birds in a Vermont mental Station. New Haven, CT. (USA) hardwood forest. Ecography 18, 97-108. Paton, P.N'.C., 1994. The effect of edge on avian nesting success: Litvaitis, J.A., 1993. Response of early-successional vertebrates to how strong is the evidence? Conserv. Biol. 8, 17-26. historic changes in land use. Conserv. Biol. 7. 866-873. Patterson. W.A., Backman, A.E., 1988. Fire and disease history of Litvaitis, J.A.. 2001. Importance of early-successional habitats to forests. In: Huntley, B., Webb, T. (Eds.), Vegetation History. mammals in eastern forests. Wildl. Soc. Bull. 29, 466-473. Kluwer Academic Publishers, Dordrecht, The Netherlands. Lorimer, C.G., 2001. Historical and ecological roles of disturbance pp. 603-632. in eastern North American forests: 9,000 years of change. Patterson. W.A., Sassaman, K.E., 1988. Indian fires in the Wildl. Soc. Bull. 29, 425439. prehistory of New England. In: Nicholas, G.P. (Ed.), Holocene Lorimer. C.G.. White, A.S.. 2003. Scale and frequency of natural Human Ecology in Northeastern North America. Plenum Press. disturbances in the northeastern United States: implications for New York. pp. 107-135. early-successional forest habitat and regional age distributions. Perala, D.A.. 1977. Manager's Handbook for Aspen in the North- For. Ecol. Manage. 185.41-64. Central States. General Technical Report NC-36. USDA Forest Major, R.E., Kendal, C.E., 1996. The contribution of artificial nest Service. NC Forest Experimental Station. St. Paul. MN. experiments to understanding avian reproductive success: a Raymond K.S., Servello, F.A., Griffith, B., Eschholz. W.E., 1996. review of methods and conclusions. Ibis 138, 298-307. Winter foraging ecology of moose on glyphosate-treated Manson, R.H., Ostfeld. R.S., Canham, D.D., 2001. Lon,*-tern clearcuts in Maine. J. Wildl. Manage. 60, 753-763. effects of rodent herbivores on tree invasion dynamics along Rudnicky. T.C., Hunter Jr., M.L., 1993a. Reversing the fragmenta- forest-field edges. Ecology 82, 3320-3329. tion perspective: effects of clearcut size on bird species richness Marquis. D.A.. 1967. Clearcutting in northern hardwoods: results in Maine. Ecol. Appl. 3, 357-366. after 30 years. Research Paper NE-85. USDA Forest Service. Rudnicky. T.C.. Hunter Jr., M.L., 1993b. Avian nest predation in Martin. C.W.. Hornbeck. J.W.. 1989. Revegetation after strip clearcuts. forests, and edges in a forest-dominated landscape. J. cuttings and block clearcutting in northern hardwoods: a 10- Wildl. Manage. 53, 358-364. year history. Research Paper NE-625. USDA Forest Service. Santillo. D.J.. Brown, P.W.. Leslie, D.M., 1989. Response of Martin, A.C., Zim. H.S., Nelson. A.L.. 1951. American Wildlife songbirds to glyphosate-induced habitat changes on clearcuts. and Plants. McGraw-Hill, New York. J. Wildl. Manage. 53. 64-71. Maurer. B.A.. McArthur, L.B.. Whitmore. R.C., 1981. Effects of Seymour. R.S., Hunter Jr., M.L.. 1992. New forestry in eastern logging on guild structure of a forest bird community in West spruce-fir forests: principles and applications to Maine. Virginia. Am. Birds 35, 11-13. Miscellaneous Publication 716. ME Agricultural Experimental May. P.G., 1982. Secondary succession and breeding bird Station. Orono, ME. community structure: patterns of resource utilization. Oecolo- Seymour. R.S.. Hunter Jr., M.L., 1999. Principles of ecological gia 55, 208-216. foresw. In: Hunter Jr., M.L. (Ed.), Maintaining Biodiversity in McAuley. D.G.. Longcore, J.R.. Sepik, G.F., Pendleton. G.W., Forest Ecosystems. Cambridge University Press. Cambridge. 1996. Habitat characteristics of American woodcock nest UK, pp. 22-61. sites on a managed area in Maine. J. Wildl. Manage. 60, Seymour. R.S., Hannah, P.R., Grace, J.R., Marquis, D.A.. 1986. 138-148. Siviculmre: the next 30 years, the past 30 years. Part IV. The McClure, J.W., Lee, T.D.. 1993. Small-scale disturbance in a Northeast. J. For. 84, 31-38. northern hardwoods forest: effects on tree species abundance Seymour. R.S.. White, AS., DeMaynadier, P.G., 2002. Natural and distribution. Can. J. For. Res. 23. 1347-1360. disturbance regimes in northeasten North America-valuating McClure, J.W.. Lee. T.D., Leak, W.B.. 2000. Gap capture in silvicultural systems using natural scales and frequencies. For. northern hardwoods: patterns of establishment and height Ecol. Manage. 155, 357-367. growth in four species. For:Ecol. Manage. 127, 181-189. Shugart Jr.. H.H., James. D.. 1973. Ecological succession of McKinnon. F.S.. Hyde. G.R., Cline, A.C., 1935. Cut-over old field breeding bird population in northwestern Arkansas. Auk 90, pine lands in central New England. Harvard For. Bull. 18. 62-77. Mills. B.B.. 1961. The Beddington bum stages a comeback. Smith, D.B.. Larson, B.C., Kelty, M.J.. Ashton. P.M.S., 1997. The Research Note 117. USDA Forest Service. Practice of Silviculture: Applied Forest Ecology, ninth ed. Noss. R.F.. LaRoe 111, E.T., Scott, J.M., 1995. Endangered Wiley. New York. ecosystems of the United States: a preliminary assessment of Stephens. G.R., Ward, J.S., 1992. Sixty years of natural change in loss and degradation. Biological Report 28. National Biological unmanaged mixed hardwood forests. Bulliten 902. CT Service. Washington. DC. Agricultural Experimental Station, New Haven. CT. K.M. DeGraqf: M. Yamasaki / Forest Ecologv arld Martagen~rnt185 (2002) 179-191 191

Suthers, H.B.. Bickal, J.M.. Rodewald. P.G.. 7000. Use of VanderHaegen. W.M.. DeGraaf. R.M.. 1996. Predation on artificial successional habitat and fruit resources by songbirds during nests in forested riparian buffer strips. J. Wild. Manage. 60. autumn migration in central New Jersey. Wilson Bull. 112. 547-550. 249-260. Vega Rivera. J.H.. Rappole. J.H.. McShea. J.H.. Has, C.A.. 1998. Taylor. C.M.. Taylor. W.E., 1979. Birds of upland openings. In: Wood thrush post fledgling movements and habitat use in DeGraaf, R.M.. Evans. K.E. (Eds.). Management of North northern Virginia. Condor 100. 69-78. Central and Northeastern Forests for Nongame Birds. General Veit. R.R.. Petersen. W.R.. 1993. Birds of Massachusetts. Technical Report NC-51. USDA Forest Service. pp. 189- Massachusetts Audubon Society. Lincoln. MA. 197. Villafuerte. R., Litvaitis, J.A.. Smith, D.E.. 1997. Physiological Thompson. C.F.. 1977. Experimental removal and replacement of responses by lagomorphs to resource limitations imposed by territorial male yellow-breasted chats. Auk 94. 107-1 13. habitat fragmentation: implications to condition-sensitive pre- Thompson 111. F.R.. DeGraaf. R.M., 2001. Consen-ation ap- dation. Can. J. Zool. 75. 148-151. proaches for woody. early-successional communities in the Webb, W.L.. Behrend, D.F., Saisorn. B., 1977. Effect of logging on eastern United States. Wildl. Soc. Bull. 29. 483494. songbird populations in a northern hardwood forest. Wildl. Thompson 111. F.R., Dijak. W.D.. Kulowiec. T.G.. Hamilton. D.A.. Monogr. 55, 1-35. 1992. Breeding bird populations in Missouri Ozark forests with Whitney. G.G.. 1994. From Coastal Wilderness to Fruited Plain: a and without clearcutting. J. Wild Manage. 56. 23-30. History of Environmental Change in Temperate North America Thompson 111. F.R.. Robinson. S.K., Whitehead. D.R.. Brawn. J.D.. 1500 to the Present. Cambridge University Press. Cambridge, UK. 1996. Management of central hardwood landscapes for the Witham. J.W., Hunter Jr.. M.L.. 1992. Population trends of conservation of migratory birds. In: Thompson 111. F.R. (Ed.). neotropical migrant landbirds in northern coastal New England. Management of Midwestern Landscapes for the Conservation In: Hagan 111, J.M.. Johnston. D.W. (Eds.), Ecology and of Neotropical Migratory Birds. General Technical Report NC- Conservation of Neotropical Migrant Landbirds. Smithsonian 187. USDA Forest Service, NC Forest Experimental Station, St. Institute Press. Washington. DC, pp. 85-95. Paul. MN. pp. 117-143. Yamasaki. M., McLellan, T.M., DeGraaf. R.M., Costello, C.A., Trani, M.K.. Brooks. R.T.. Schmidt, T.L., Rudis. V.A.. Gabbard, 2000. Effects of land-use and management practices on the C.M., 2001. Panerns and trends of early-successional forests in presence of brown-headed cowbirds in the White Mountains of the eastern United States. Wildl. Soc. Bull. 29, 413124. New Hampshire and Maine. In: Smith, J.N.M Cook. T.L. Tryon. H.H.. 1945. The management of young volunteer hardwood Rothstein. S.I. Robinson, S.K. Sealy. S.G. (Eds.). Ecology and stands. Black Rock Forest Bull. 13. Cornwall-on-the-Hudson. Management of Cowbirds and their Hosts. University of Texas NY. Press, Austin. TX, pp. 31 1-319. Tubbs. C.H.. Verme. L.J., 1972. How to create wildlife openings in Zimmerman. J.L.. 1988. Breeding season habitat selection by the northern hardwoods. Unnumbered Pamphlet. USD.A Forest Henslow's sparrow (Anm~odranlus henslowii) in Kansas. Service, NC Forest Experimental Station. St. Paul. hlN. Wilson Bull. 100. 17-24.